Page 1

Newsletter of the Center for Proton Therapy  ::  Paul Scherrer Institut  ::  August 2017  ::  # 12 :: page 1


Center for Proton Therapy  ::  Paul Scherrer Institut  ::  # 12_8/2017

Dear Reader

If there is one category of patients who will benefit from protons,

vanced proton delivery techniques that include motion-mitigation

it is surely children, adolescents and young adults (AYA). In this strategies such as re-scanning, with or without gating. The former Welcome to this summer edition of SpotOn+. As of July 17th, the

edition, we report our results of proton therapy for EWING sarcomas has been now applied routinely in patients for which tumor and/

clinical program at PSI has successfully resumed as planned when

for these challenging patients. With a follow-up of approximately or OAR’s motion is a clinical concern, with the first patient treated

the COMET cyclotron went back on line two weeks prior to this date.

50 months, only 16% of children/AYA experienced a local failure.

this summer. The end-to-end testing showed that spatial displace-

We are now having a full patient throughput with anesthesia patients Importantly only 2 (7%) patients experienced significant (i.e. grade

ment of the measured dose was acceptable with a standard devi-

mostly treated at Gantry 2. This choice has been self-evident in the 3) late toxicity due to the decrease in radiation delivery to tissues ation of 0.0–0.3 mm. We are excited in the prospect of introducing light of the speed of the radiation delivery with this treatment unit in vicinity of the target volume. Despite the numerous negative optical tracking in addition to the above-mentioned motion-mitiwhich has consequently decreased the sedation time for children. prognostic factors presented by a substantial number of patients, gation strategies. As of 2017, we are the only proton therapy center Cases can be referred to PSI by contacting the head of the clinical the survivorship at five years (>80%) was excellent. The second worldwide proposing combined motion-mitigation strategies, inteam (Dr. Marc Walser; marc.walser@psi.ch) or myself (damien.we-

article of this edition deals with advanced MRI techniques (i.e. cluding rescanning techniques, to alleviate the interplay-effect

ber@psi.ch). The next big task on our list is the clinical commission-

functional imaging) used for proton therapy planning purposes, inherent to PBS delivery to moving target volumes.

ing of our new treatment Unit (Gantry 3) that will start this late summer.

with or without a SIB delivery paradigm. The reader has to go be- That said, I would like to thank you for reading this Newsletter and

On the picture above you can see the coupling point of the beam line yond the rather tedious list of acronyms (DWI, DTI, DSC, ASL), to wish you a good start after the summer recess. Stay tuned on (on the left hand side) with the G3. It is foreseen that the first patients acknowledge the interest of imaging for radiation delivery in SpotOn+ to have the latest clinical/research reports on our center. should be treated by the end of this year. A dedicated SpotOn+ edition general and protons in particular. Our group has shown that ‘func-

 Yours sincerely,

will be proposed soon to our readers and a symposium will be sched-

tional sparing’ was possible using dedicated MRI sequences that

uled in Q2 2018 at PSI. Additional information will be given to the

may benefit ultimately a selected number of brain tumor patients.

Prof. Dr.med. Damien Charles Weber

radiation oncology community in a not too distant future.

Finally, the last article details the clinical commissioning of ad-

Chairman of CPT, Paul Scherrer Institute

Newsletter of the Center for Proton Therapy  ::  Paul Scherrer Institut  ::  August 2017  ::  # 12 :: page 2

Radio-Oncology News Pencil-beam scanned protons for the treatment of patients with Ewing sarcoma Introduction

proton radiation therapy is that dose

with propofol under spontaneous

distribution may be possibly optimized breathing. Size of the tumour ranged Ewing sarcoma (EWS) is a highly malig-

and the production of neutrons de-

from 1.7 to 24 cm (median: 6.7 cm).

nant small round-cell tumour of the

creased, thus decreasing the probabil-

Most common primary site was axial/

bone and/or soft tissue, most com- ity of radiation-induced secondary pelvic (n=27; 71%). Four patients (11%) monly found in male adolescents and cancers. To the best of our knowledge presented with metastasis at dia­ young adults (AYA) with a highest inci- this analysis presents the first results gnosis. Twenty (53%) patients had dence in the 10–15 years range. The

on the long-term outcome of children

chemo-PT only. Median follow up was

therapeutic strategy for these challeng-

and AYA treated with PBS proton ther-

49.6 months (range, 9.2–131.7). Local

ing tumours consists on multimodal apy (PT). As such, we have evaluated control (LC), distant metastasis-free treatment involving chemotherapy and the outcome of these patients and have

Figure 1: Treatment plan of a 18-year old male patient, treated for a Ewing sarcoma located in the sacrum to a total dose of 66.5 GyRBE.

tuarial rate of local control (LC), distant


survival (DMFS), toxicity-free survival metastasis-free survival (DMFS) and

local therapy, the latter consisting of assessed the major prognostic factors and overall survival (OS) were deter-

overall survival (OS) were 81.5%, 76.4%

These preliminary data suggests that

mined from the first day of PT. Univar-

and 83.0%, respectively. All local re-

the outcomes of children and AYA with

sulting in a 5-year survival of approxi-

iate analyses were performed to iden-

currences occurred in field and in pa- EWS were good and PT was well toler-

mately 60–75% for non-metastatic Materials and Methods EWS. The advantage of pencil-beam

tify prognostic factors.

tients with non-extremity primaries. Six ated with very few late adverse events.

scanning (PBS) over passive-scattering To identify the study cohort, we se-


surgery and/or radiotherapy (RT), re-

for this locally invasive tumour.

lected all paediatric and adolescent/

Figure 2: Local tumor control as a function of age of EWS patients treated with PT.

< 10 years

patients died, all of tumour progres- The local and distant tumour control for sion. Age < 10 years was a favourable older patients with large pre-PT tumour factor for LC (P=0.05) and OS (P=0.05) volumes remains problematic in these

young adult (AYA) patients aged ≤ 39

With a median follow up of 49.6 months of borderline significance, but was sig-

years with at least 9 months of fol-

and 50.5 months for the surviving pa- nificant for DMFS (P=0.003). Tumour The results were recently published

low-up (FU). We excluded 3 cases with tients, 6 (15.8%) patients experienced volume < 200 ml was a significant prog-

challenging patients. (Weber al. 2017) and will be presented

FU < 9 months and another case aged local failures after a median time of nostic factor for DMFS (P=0.03), but at the 49th Congress of the International ≥ 10 years

> 39 years. Thirty-eight patients (me-

22.4 months. Seven (18.4%) patients not for OS (P=0.07). Metastasis at di-

Society of Paediatric Oncology (SIOP)

dian age at diagnosis: 9.9 years, range: developed distant metastasis 2.5-56.1 agnosis was a strong predictor of local in October in Washington. 0.4 – 38.9 years) form the basis of this months (median, 19.2) after PT. Two report. A total of 24 male and 14 female

patients received a median dose of local and distant failure, 2 other paP=0.05


mucositis. Only 2 (6.9%) grade 3 late Weber et al. Pencil beam scanned

54.9 GyRBE (range: 45.0 – 69.6 GyRBE). tients presented with sequential local toxicities were observed. The 5-year

protons for the treatment of

and distant failure and 3 patients with

actuarial rate of grade 3 toxicity-free

patients with Ewing sarcoma

survival was 90.9%.


Thirteen (34%) patients were anesthe-

tized during PT. Children were sedated distant metastasis only. The 5-year acFigure 2. Local tumor control as a function of age of EWS patients treated with PT

failure (P=0.003). Acute toxicity was

patients presented with concomitant limited to grade 1–2 skin erythema or

Newsletter of the Center for Proton Therapy  ::  Paul Scherrer Institut  ::  August 2017  ::  # 12 :: page 3

Medical-Physics News Shaping proton therapy dose with DTI and DSC MRI data: functional SIB and avoidance proof of concept study External beam radiation therapy is,

stantially affect patient quality of life.

together with surgery and chemother-

Current treatment planning typically cytoma (WHO II) patient (Figure 1), four

to the rCBV. For an example oligoastro-

apy, a standard first (radical) or second ignores functional structures, such as treatment plans have been calculated. line therapy for many brain cancer in-

white matter neural networks or hy-

i) A uniform dose PBS proton plan, ii) a

dications. In particular, ions (e.g. pro-

poxic/aggressive sub-areas of tumours.

Simultaneous Integrated Boost (SIB)

tons and carbon) are attractive because

However advanced MR imaging tech-

plan based on a rCBV, iii) a uniform plan

of their ability to control dose deposi-

niques, such as diffusion weighted modified with DTI_CST guided functional

tion at a precise location in space due

imaging (DWI), diffusion tensor imag-

sparing and iv) both SIB and DTI_CST

to their Bragg peak characteristic, al-

ing (DTI), perfusion with Dynamic Sus-

functional sparing. Dose constraints for

lowing improved sparing of organs at ceptibility Contrast (DSC) or Arterial the DTI_CST plans were defined either risk. In addition, proton pencil beams Spin Labelling (ASL) MRI and MR-Spec-

on the tract structures as a whole volume

used in combination with Intensity troscopy (MRS) provide the possibility or by splitting them into upper, middle

reduced by 5% and 2% in the brainstem

and lower portions, as the middle por-

and cochlea respectively. This was how-

Modulated Proton Therapy (IMPT) allow

to more precisely characterize brain

for flexible planning of dose and sharp

lesions and functional regions to aid tion overlapped with the PTV.

ever at the cost of a reduced V90 to the

dose gradients even in the most com-

the treatment planning process.

CTV (91.3% vs 100%). With different

plex geometries. Nonetheless, radia-

In this work, we investigate the potential Figure 2. For the rCBV-SIB plan (ii), mean constraints for the different DTI_CST

tion-induced damage to healthy tis-

of incorporating cortico-spinal tract and max doses in the cochlea were re-

sues in the form of early, delayed or

structures (DTI_CST), determined using

late neurocognitive toxicity can sub-

DTI, and relative Cerebral Blood Volume the uniform plan, max dose to the brain-

Resulting plans and DVHs are shown in

portions, mean and max dose to CST_

duced by 3% and 5% respectively c.f. centre could be reduced to 52 and 56%

Figure 2: Plan Comparison. (a) clinical SFUD plan; (b) IMPT rCBV-SIB plan and (f) DVH comparison SFUD vs IMPT rCBV-SIB; (c) IMPT DTI_CST sparing plan and (g) DVH comparison SFUD vs IMPT DTI_CST; (d) IMPT with rCBV-SIB and DTI_CST sparing plan and (h) DVH com­ parison SFUD vs IMPT with rCBV-SIB and DTI-CST sparing; (e) Structures used during plan comparison.

respectively. Similar results were found

(rCBV), thought to

stem by 3% and dose to the central and for the SIB_DTI plan (iv).

correlate with tu-

lower DTI_CST by 2%. Using an alterna-

In conclusion, it has been shown that This project was presented at the

mour aggressive-

tive field arrangement, the CST-centre

IMPT plans can be generated to account BiGART conference in Aarhus, Denmark

ness and identified could be better spared (mean dose for constraints on MRI defined funcusing T2* DSC perfu-

in June this year.

-5%), but at the cost of increased mean/ tional volumes. Although this is only a

sion imaging, into max dose to the cochlea (3% and 5% first proof of concept, we think that treatment planning

respectively). For the DTI_CST sparing

introducing functional information into For any further information,

for IMPT. The goal is plan (iii), CST_centre mean (61% vs the plan and along the treatment please refer to CPT, Figure 1: Case presentation. (a) T2 FLAIR MRI with overlaid GTV and PTV contours. (b)T2 FLAIR MRI overlaid with volume with elevated rCBV (correlating with aggressiveness. (c) DTI tractography, note how the ipsilateral DTI_CST is next to the tumour area

to spare dose to the

103%) and maximum (70% vs 103%)

DTI_CST structures doses were substantially reduced c.f.

course will help to exploit the flexibility Dr. Marta Peroni of pencil beam scanned proton therapy Tel. +41 56 310 4037

whilst boosting dose the uniform plan (i), and max dose was further.


Newsletter of the Center for Proton Therapy  ::  Paul Scherrer Institut  ::  August 2017  ::  # 12 :: page 4

Figure 3: An example of gamma analysis for a patient treatment field. The gamma score improves with increased number of rescans and dose distribution remains robust also in case of fluctuations of amplitude and period.

Medical-Physics News Clinical commissioning of rescanned pencil beam scanning treatments of moving targets In the spring 2017, CPT PSI has performed com-

control system of Gantry 2 for the given rescanning ever to 97.8% (1SD=2.9%) in the presence of mod-

missioning of the clinical procedure for the treat-

regime (figure 1). In the 4DDC, 8 rescans were found erate motion fluctuations, and was not clinically

ment of moving targets with pencil-beam scanning

to provide full and homogeneous CTV coverage for

on Gantry 2. This included treatment planning with all cases, independent of starting phase (fluctua4D dose calculation (4DDC), patient-specific ver-

platform, and end-to-end testing using a dynamic anthropomorphic phantom.

GS=76.9%) (figure 3).

tions of V95% and D5%–95% under 2%).

ifications extended by measurements with a 2D ionization chamber array mounted on a moving

acceptable for amplitudes >10mm (average

Dosimetric verification

End-to-end test The whole clinical treatment workflow, including

Each rescanned field was measured at proximal and 3-DOF patient positioning based on image registradistal depth using a 2D-chamber array as stationary tion of the averaged 4DCT with a slow, pre-treatment For any further information, please refer to CPT,

Treatment planning approach and 4D dose calculation (4DDC)

and also moving (figure 2) under different motion in-room CT, was verified using an anthropomorphic Dr. Jan Hrbacek scenarios:

breathing phantom and radiochromic films posi- Tel. +41 56 310 37 36, jan.hrbacek@psi.ch

• nominal (8mm motion),

tioned at 2 planes in the tumour. Spatial displace-

Single field uniform dose (SFUD) plans were opti-

• random fluctuations in amplitude and period

ment of measured dose distributions were 1.4mm

mized on the mid-ventilation phase extracted from

• scaled in amplitude to exceed the clinical inclu-

(1SD=0.3mm) in the left-right direction and 1.5mm

a 4DCT dataset of each thoracic patient, with primary

sion criteria (>10mm).

anatomical target motion below 8 mm in superior-in-

Gamma score (GS) (3% dose difference /3 mm

ferior direction, delivering V95%=100% to PTV.

distance-to-agreement) was 100% for all static meas-

(1SD=0.0mm) in the cranio-caudal direction. Dose inhomogeneity in the CTV was <9%.

Static plans created for the mid-ventilation phase

urements and >98.5% for all fields tested with Conclusion and outlook were subject to 4DDC, which simulates the interplay nominal patient motion. Average GS reduced howbetween motion of anatomy derived from 4DCT and

It was theoretically and experimentally verified that

time structure of delivering pencil beam sequence,

rescanning can be safely applied clinically to miti-

taking into account handling of all spots by treatment

gate motion with irregularities in moving-target treatments for amplitudes ≤ 8mm. Optical tracking system is currently under integration at Gantry 2 to allow treatment of targets moving more than 8 mm

Chairman Prof. Damien C. Weber Chief Medical Physicist Prof. Tony Lomax Design and Layout Monika Blétry


ence of the particle therapy co-operative group (PTCOG) mid of May this year in Yokohama, Japan.

Villigen PSI, August 2017

be commissioned for the clinical use by the end of this year. This work was presented at the 56th annual confer-

Figure 1: Schematic diagram of the 4D dose calculation.

Editor Dr. Ulrike Kliebsch

Center for Proton Therapy CH-5232 Villigen PSI protonentherapie@psi.ch www.protonentherapie.ch Tel. +41 56 310 35 24 Fax +41 56 310 35 15

in the regime of gated rescanning. This system will

Figure 2: A 2D ionization chamber array mounted on a moving platform placed under a “bridge” which holds a water phantom.


Profile for Paul Scherrer Institut

PSI Protontherapy Newsletter Nr. 12 (08/2017)  

PSI Protontherapy Newsletter Nr. 12 (08/2017)